Wafer rotary switch repair?

[NZ5N]

Hello,

Am working to restore a Gonset Communicator II.  The good news is that today I found the source of the problem.  The bad news is that the problem is the T/R switch, a 2-wafer 8PDT rotary, which I assume will be difficult to repair or replace.  So many wires attached to it, will probably take at least an hour to unsolder everything and pull the switch out.

It looks like the switch is not burned or broken.  The problem seems to be that, on both wafers, the little metal ring that moves when the switch is turned seems to be out of place.  The small extensions on the rings that are supposed to make contact with the terminals move back and forth when the switch is turned, but they are not even close to touching the contacts.

Is this a common problem with old switches?  Any thoughts on the best way to proceed?

73, Bill NZ5N   

[W6EM]

Bill-

On some of the older wafer switches, a small keeper just in front of the bushing and under the knob was used to hold the shaft in position.  If the rotating disks have both moved beyond the contacts, it could be caused by shaft slippage into the radio.

To check, remove the knob and see if the shaft moves in and out.

To fix, you'll need to find a replacement keeper ring to fit the slot on the shaft when you realign the rotary disks.

73,

Lee
W6EM
 

[AC5UP]

I have never seen that particular switch, but that hasn't stopped me in the past from offering advice as I have seen plenty of rotary switches... 

Some switches have an index bar immediately on top of the detent mechanism, front plate behind the first mounting nut, that sets either the home position or number of positions within an adjustable arc. If there is an index arm make sure it hasn't worked loose enough to have slipped one click away from where it needs to be. Another possibility is the switch needs to be lined up with the front panel. Look for an index pin on the switch and matching hole in the panel. Sometimes a little bit of slack in the mounting nut can allow the index pin to slide out of the index hole.

Previous owner (?) may have had a heavy hand and the rotating wafer is now worn on the shaft... Most shafts have flats on two sides and the wafer hole is cut to match. If that hole has enlarged to the point where the wafer can wander around the shaft you might be able to replace only the switch rotor, but a switch transplant is more likely. If you have a digital camera be sure to take plenty of pictures before you change anything........

[NZ5N]

Thanks for the replies.

Lee, yes, my switch has a keeper as you describe, but it seems to be on there good and tight, the shaft is not moving at all.

Nelson, great idea about taking pictures, there are more than 24 wires attached to the switch (8PDT, some positions have more than one wire attached).  I won't have a good view of the innards of the switch until I disconnect some of the wires, do not now see any loose pin or wear in the wafer hole.  In fact the switch looks good from the outside, perhaps the problem will be obvious once I get inside.

If all else fails, there are some nice looking Russian military switches on eBay that might make a decent replacement.

73, Bill NZ5N

[NZ5N]

Now this is odd.  I played with the radio some more and discovered that the knob can be clicked several times so that it turns 360 degrees.  It is only supposed to have 2 positions, one for TX and the other for RX.  Getting some strange meter readings across the contacts, something does not seem right.

[KA5N]

Sounds like someone has manhandled the switch or maybe removed stops from the index.
It used to be simple to find replacement switches or use parts kits where you could assemble
your own switches with various positions and types of switching (make before break, progressive shorting etc.).  Unfortunately those days are gone and the chance of finding
nos parts kit switches is nil. 
About your only hope is to find a parts Gonset or to find old rotary switches at a hamfest
and take good parts off of old switches. 
It helps to have a lot of experience with switches and it takes awhile for a nubie to learn the
ins and outs.
Good Luck
Allen

[N3QE]

Now this is odd.  I played with the radio some more and discovered that the knob can be clicked several times so that it turns 360 degrees.  It is only supposed to have 2 positions, one for TX and the other for RX.  Getting some strange meter readings across the contacts, something does not seem right.

This is not supposed to be hard :-).

At the front of the radio is the detent mechanism which has a limit prong or tab. Someone has applied too much force and busted off the prong/tab, or disassembled and reassembled the mechanism improperly, or the lock nuts just loosened up and this happens.

If the knob has been forced and the limit prong popped off, or if the nuts came loose and the whole mechanism rotated, it is likely that other things are out of alignment. Again, this is not hard. Take apart a couple other vintage rotary switches in your junkbox and you'll probably find any missing parts that you need. Worst case a radio shack rotary switch (275-1386) might be disassembled to get equivalent limit/alignment functionality.

[NZ5N]

Thanks much for the replies.  Turns out that the T/R switch is OK.  It can be clicked several times, it will turn a full 360 degrees.  Don't know if this is a defect or if all Communicator switches do this, but as long as it is left in the correct position it works as it should, I checked all the contacts with a meter.

So the good news is that the switch is OK (or at least useable).  The bad news is that this means I still do not have a cure for my original problem.  This rig receives fine but no output on TX, the magic eye does not move regardless of where the tuning capacitors are turned.  All tubes check out OK on a tube tester.  Power supply is OK, a prior owner replaced the rectifier tubes with diodes and the voltage is higher than specs (380v rather than 300v).

The manual has a schematic with numerous test points and a handy voltage chart listing what the voltage should be at each point.  However, the only voltage chart I have is for the Communicator I, and my rig is a Communicator II.  The layout and test points are the same but many component values are different, but the spec voltages may be different for my rig.  Plate voltage is somewhat higher on each tube due to the power supply, screen voltages are generally lower but in the ballpark.  The only significant differences I see are the grid bias voltages in the 12AV7 driver and the 2E26 final.  The chart says they should be -45v but I am reading -8v.

Any thoughts on where I should go from here?

73, Bill NZ5N

[KB4QAA]

Do the hard work first.

Fix the stops on the T/R switch.  If you don't nothing but bad can happen when you absent mindedly turn it too far, with flashes and smoke to follow. 

[NZ5N]

Yes, I'll definitely fix the stops if I can get this thing working.

I can hear the radio transmitting on a nearby receiver, on the appropriate frequency with each of the 2 crystals. I peaked each of the tuning capacitors for maximum S-meter reading on the receiver and am now showing a bit of power output on an external meter, about 1/2 watt out. So it seems the 6CL6 oscillator section is OK and the problem must be connected with the 12AV7 driver or the 2E26 final. These are the tubes with the odd bias readings.

The low bias does seem to be the primary suspect at this point. I do not quite understand how the negative bias voltage is generated. There is not a separate bias supply. Each of the 2 grids at issue has connected to it a
coupling capacitor to the plate of the previous tube, and an RF choke. On the
other side of the choke there is a 22k resistor. The other side of the resistor
goes to ground. The test point is the spot where the RF choke connects to the
22k resistor.

I'm not clear to me why low grid bias would cause such low output, but that seems to be the best clue I have.

If you or anyone else is interested in seeing the schematic, the one for my Comm
II (without parts list and voltage chart) is at:
http://bama.edebris.com/download/gonset ... om2sch.pdf

The schematic for the Comm I ((with parts list and voltage chart) is at:
http://bama.edebris.com/download/gonset ... a/gcom.pdf

If anyone knows where to find the Comm II parts list and voltage chart, please
let me know.

Would appreciate any further ideas.

73, Bill NZ5N

[KA5N]

The bias source is the voltage developed across the grid leak resistor and is called
"Grid Leak Bias" and no "bias supply" is required. 

You don't have enough drive to develope the correct power and hence grid bias.  With a half watt output you have verified this. 

It may be that some of the tubes are weak or the screen voltages are too low or who knows
what?  You cannot properly trouble shoot equipment if you don't understand how it works.
Understanding comes from study and experience.

Good Luck
Allen

[AC5UP]

Understanding comes from study and experience.

...plus good documentation with the time it takes to help you understand how a piece works.

With boatankers it's not unusual to substitute a schematic for a similar model or select a near-identical part when the original becomes unobtanium, but to troubleshoot a radio without a good understanding of how the circuits work together is almost a guarantee of frustration. If all you can find is a schematic for the II it might be worthwhile looking at the docs for the III or IV on BAMA. Different model, different circuitry, but it's rare for a manufacturer to fully re-design a piece from one model to the next.

[NZ5N]

Thanks for the replies. A solution may be near, found a couple of issues:
* someone loaned me a tube tester and every tube is fine, except the 12AV7 is reading low on both triodes, supposed to get a minimum of 21 on the meter and only getting 10 on one side and 14 on the other. Hopefully someone will have one for sale at the Orlando hamfest this weekend.
* the 22k resistor in the 2E26 grid circuit is only reading 400 ohms. That would account for the low bias on the 2E26 grid, although the 22k resistor on the 12AV7 is reading the correct value.

Also someone has promised to email me a parts list and voltage chart for the Comm II.

73, Bill NZ5N

[NZ5N]

This started as a promising evening but ended with no real progress.  It turned out that the 22k resistor was OK, it was just that there is a 390 ohm resistor in parallel with it when the T/R switch is in Receive.  I obtained the parts list and voltage chart for the Comm 2 from a helpful ham.  The differences from the Comm I are substantial.  Next, a friend gave me three 12AV7s to try.  I was confident the rig would be transmitting at higher power before long.

But it was not to be.  Two of the 12AV7s tested lower than the one in the rig.  Each of the two triodes in the tube is supposed to show a minimum of 21 on the tester, mine are reading 14 and 10, the best of the 3 borrowed ones reads 22 and 16.  With the borrowed tube, bias on both the 12AV7 and the 2E26 went to about -30v on both tubes, much better than before.  But output (even after retuning) actually went down a bit, from .5 watts to .38 watts.  Very frustrating.  Probably bias would be closer to specs with a better tube, but that does not seem to be making any difference.  Why would output go down rather than up?  Don't get it.

After noticing that the cathode voltage on the 12AV7 was low (2.6v instead of 4.0v), I checked the 100 ohm resistor R6 and it measured just 38 ohms.  Perhaps that is significant.  Still have not checked the coupling capacitors, not sure how to do that except by replacing them.

As mentioned yesterday, a previous owner replaced the rectifier tubes with diodes, which increased B+ by about 25%.  I have seen 2E26 projects using more than 500v, so not sure how important this is, the guy who sold it to me said that it worked for him for years.

Here are the pertinent voltage chart/measured voltages (T/R switch on Transmit except as noted):
A 1.6/1.4 6CL6 cathode
B 135/135 6CL6 screen
C 255/280 6CL6 and 12AV7 anodes
D 4.0/2.6 12AV7 cathodes
E -90/-33 12AV7 grid (triode 2)
F -55/-29 2E26 grid
G 165/155 2E26 screen
JJ (RX) 300/390 B+ from power supply
JJ (TX) 255/305
KK -100/-95 12AV7 (triode 1)

Any further thoughts?

73, Bill

[NZ5N]

Better results tonight.  Looks like the primary culprit was that 12AV7.  Someone loaned me another one, this one tested good in the tester, and when I put it in the output increased to 4.5 watts.  Now I'm just debating whether to "declare victory" and leave it alone (after fixing the T/R switch stop), or to try tweaking it further to hopefully get the full 6 watts that should be there.  I'm still concerned about the 12AV7 cathode resistor R6, which is supposed to be 100 ohms but is reading just 38 ohms.  What effect would replacing that have?   

Also hooked up a D-104 mike and it sounded fine.  Had to set the switch on the back to Carbon rather than Crystal, not sure why, as the D-104 is a crystal mike.  No modulation at all when set to Crystal.

Thanks to everyone for the help!

73, Bill